Methodology for the Development of Reflexive-Active Systems of Artificial Heterogeneous Intelligent Agents

When developing automated systems for solving practical problems, it is necessary to take into account their weak formalization, complex structure, heterogeneity, network nature of conditions and goals, significant opacity (uncertainty) and dynamism. Under these conditions, the developer needs a tool that ensures the organization of productive activity into a holistic system with clearly defined characteristics, logical structure and implementation process. For this purpose, this paper proposes methodology for constructing reflexive-active systems of artificial heterogeneous intelligent agents that act as a computer model of a team of specialists solving practical problems without their significant simplification, taking into account the listed characteristics.